Artifact reduction of dental implants on high resolution MR imaging
Tim Hilgenfeld1, Alexander Heil1, Sebastian Schwindling2, David Grodzki3, Mathias Nittka3, Daniel Gareis4, Peter Rammelsberg2, Martin Bendszus1, Sabine Heiland1, and Marcel Prager1

1Division of Neuroradiology, University Heidelberg, Heidelberg, Germany, 2Division of Prosthodontics, University Heidelberg, Heidelberg, Germany, 3Siemens Healthcare GmbH, Erlangen, Germany, 4NORAS MRI products GmbH, Höchberg, Germany


Dental MRI is a new technique which is often impaired by artifacts due to metallic dental implants. Several MRI sequences were developed to reduce susceptibility artifacts (e.g. for orthopaedic implants). Here, we for the first time systematically evaluated MR sequences for artifact reduction in dental implants. Smallest artifact volume was measured for 2D-TSE sequences. Since imaging of dental structures benefit from high resolution and possibility of 3D reconstructions 3D sequences are advantageous. Significant artifact reduction was noted for SPC-WARP measuring only 2.1 times artifact volume of TSE sequence instead of 4.8 times when using standard SPC sequence.


Magnetic resonance imaging (MRI) has become a standard diagnostic tool for head and neck imaging. In addition new potential dental MRI applications are in developmental stage for example in orthodontics [1, 2], endodontics [3], prosthodontics [4] and periodontology [5]. In patients undergoing dental MRI, image quality may be impaired due to susceptibility artifacts due to metallic dental implants [6]. Several MRI techniques to reduce susceptibility artifacts were developed like view angle tilting (VAT), slice-encoding metal artifact correction (SEMAC), pointwise encoding time reduction with radial acquisition (PETRA), multiple slab acquisition with VAT gradient based on a SPACE sequence (SPC-WARP), and techniques employing combinations of those like TSE-WARP (VAT and SEMAC). Up to now, these sequences were predominantly evaluated in orthopaedic or neurosurgical metallic implants [7-10]. The aim of this study was to determine the value of these techniques in terms of artifact volume reduction at different dental implants while maintaining reasonable acquisition times compared to standard sequences.


Two dental implants with different composition of crowns (CCT-T implant: cobalt, chromium and tungsten; Z-T implant: zirconium dioxide), prosthesis screws and abutments made of titanium were used. For measurement of artefact volume implants were embedded in a mixture of water and fat. Afterwards image quality was assessed in porcine head with implants placed in anterior mandible. All MR measurements were performed on a 3T MRI system (Tim-Trio; Siemens Healthcare GmbH). A 16-channel multipurpose coil (Variety; NORAS MRI products; measurement of artefact volume and imaging of porcine head) and 12-channel head coil (Siemens Healthcare GmbH; signal-to-noise ratio (SNR) measurement) were used. TSE-WARP, SPC-WARP and PETRA sequences were optimized with regards to artifact size (for example matrix size, readout bandwidth, and slice thickness). In a second step standard TSE and SPACE (SPC) sequences with imaging parameters identical to WARP sequences as much as possible were implemented for comparison. Determination of artifact volume was performed by subtraction of the true implant volume determined by water displacement from the volume measured in MRI. Semi- automatic quantification and rendering of signal loss- and pile up artifact volume were done with Amira 3D (FEI). Image quality was assessed quantitatively by calculating SNR and qualitatively by blinded read of porcine heads by two radiologists. The SNR was determined by measuring the dynamic noise [11]. 25 repetitions of each sequence were performed. Using Matlab R2015a (MathWorks Inc.) Regions of interest were automatically placed exactly at the same position in all sequences and a voxel based SNR map was calculated. Radiologists evaluated image quality of eight anatomical structures for the last molar on a scale from 1 (best visibility) to 5 (poor visibility) as published before [8].


Comparing different implants, the larger artifact volume was always noted for the CCT-T implant which was between 9.3 ± 2.4 times (TSE-WARP) and 31.6 ± 4 times (PETRA) larger than the artifact volume of the Z-T implant. For both implants no significant reduction of artifact volume was observed using the TSE-WARP sequence compared to a standard TSE sequence. In contrast, SPC-WARP reduced the artifact volume of the CCT-T implant by 56.2 ± 3 % and 33.1 ± 6.8 % for Z-T implant, respectively compared to standard SPC sequence. A significant increase in artifact volume of CCT-T implant compared to standard SPC sequence was observed for PETRA (+11.8 ± 4.2%) whereas artifact volume of Z-T implant decreased by 40.8 ± 6.3%. Significant but minor differences in SNR were noted between SPC versus SPC-WARP and TSE versus TSE-WARP, respectively. SNR of PETRA was about 69.5 ± 8.5 % smaller than for standard SPC which resulted in poor performance in qualitative image review as well (significant worse in six out of eight anatomical structures compared to standard SPC). No significant differences were observed between SPC versus SPC-WARP and TSE versus TSE-WARP, respectively. Better results (but not significant) were obtained for small structures like periodontal space or apical foramen with SPC and SPC-WARP. Cohen’s κ for interrater agreement was excellent (0.81).


There was no significant artifact reduction using TSE-WARP compared to standard TSE sequences. Artifact volume was smaller in all TSE sequences compared to SPC, SPC-WARP and PETRA sequences. However, in dental MRI acquisition of 3D sequences is advantageous for various reasons. Artifact reduction of dental implants can be achieved with SPC-WARP and PETRA technique compared to standard SPC but the effect depends on composition of dental implants. Poor image quality was obtained with PETRA. SPC and SPC-WARP performed slightly better in terms of image quality than TSE and TSE-WARP as a result of higher resolution.


We thank SIEMENS Healthcare GmbH for providing two WIP packages and Straumann Germany for providing the zirconia prosthesis screw. Furthermore the authors would like to thank Stefanie Sauer, Ph.D., pharmacist in the Department of Pharmacy Heidelberg University Hospital for her work on this project.


1. Eley, K.A., S.R. Watt-Smith, and S.J. Golding, "Black Bone" MRI: a potential non-ionizing method for three-dimensional cephalometric analysis--a preliminary feasibility study. Dento maxillo facial radiology, 2013. 42(10): p. 20130236.

2. Tymofiyeva, O., et al., Three-dimensional localization of impacted teeth using magnetic resonance imaging. Clinical oral investigations, 2010. 14(2): p. 169-176.

3. Kress, B., et al., Age- and tooth-related pulp cavity signal intensity changes in healthy teeth: a comparative magnetic resonance imaging analysis. Oral surgery, oral medicine, oral pathology, oral radiology, and endodontics, 2007. 103(1): p. 134-137.

4. Tymofiyeva, O., et al., In vivo MRI-based dental impression using an intraoral RF receiver coil. Concepts in Magnetic Resonance Part B: Magnetic Resonance Engineering, 2008. 33B.

5. Schara, R., I. Sersa, and U. Skaleric, T1 relaxation time and magnetic resonance imaging of inflamed gingival tissue. Dento maxillo facial radiology, 2009. 38(4): p. 216-223.

6. Penarrocha Diago, M., A. Boronat Lopez, and J. Lamas Pelayo, Update in dental implant periapical surgery. Med Oral Patol Oral Cir Bucal, 2006. 11(5): p. E429-32.

7. Ai, T., et al., SEMAC-VAT and MSVAT-SPACE sequence strategies for metal artifact reduction in 1.5T magnetic resonance imaging. Investigative radiology, 2012. 47(5): p. 267-276.

8. Lee, Y.H., et al., Usefulness of slice encoding for metal artifact correction (SEMAC) for reducing metallic artifacts in 3-T MRI. Magn Reson Imaging, 2013. 31(5): p. 703-6.

9. Sutter, R., et al., Reduction of metal artifacts in patients with total hip arthroplasty with slice-encoding metal artifact correction and view-angle tilting MR imaging. Radiology, 2012. 265(1): p. 204-214.

10. Cho, Z.H., D.J. Kim, and Y.K. Kim, Total inhomogeneity correction including chemical shifts and susceptibility by view angle tilting. Medical physics, 1988. 15(1): p. 7-11.

11. Dietrich, O., et al., Measurement of signal-to-noise ratios in MR images: influence of multichannel coils, parallel imaging, and reconstruction filters. Journal of magnetic resonance imaging : JMRI, 2007. 26(2): p. 375-385.


Three-dimensional volume of pile up and signal loss artifacts of CCT-T implant (A) and Z-T implant (B) in T1 weighted images using two and three-dimensional sequences with and without artifact suppression (data presented as mean ± standard deviation; *: p ≤ 0.5; **: p ≤ 0.01; n.s.: not significant)

Three-dimensional artifact rendering of CCT-T implant (first row; red: pile up artifact; blue: signal loss artifact) and coloured source images (middle row). Curved oblique sagittal reconstructions of last molar of porcine head (bottom row).

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)